Publiée le 18 juin
Mission du poste
Senior Voice AI Engineer
Brussels | Hybrid | 50% on site, 50% remote
Freelance 12-month contract)Start: July 2026
We’re looking for a Senior Voice AI Engineer to help build a real-time, client-facing Voice Bot for a major enterprise client in the financial services sector.
You’ll join a team of AI Dev Engineers and Data Scientists delivering a production-grade hybrid speech system combining open-source ASR/TTS/VAD models with vendor APIs, optimised for low latency, reliability, and conversational quality.
Roles & Responsibilities
- Build end-to-end streaming voice pipelines (audio → VAD → STT → LLM orchestration → TTS)
- Own real-time conversation behaviour: turn-taking, barge-in / interruption handling, end pointing & silence tuning
- Design resilient streaming systems (retries, backpressure, fallback strategies)
- Collaborate with Data Scientists on production-ready ML model integration
- Contribute to CI/CD, Dockerisation, testing, and deployment of ML services
- Support monitoring, evaluation, and lifecycle management of speech models
Skills & Qualifications
- 4+ years engineering experience, including 2+ years in production speech / voice AI systems
- Strong Python (Go/Rust/C++ a plus)
- Experience with streaming audio systems and real-time ML pipelines
- Familiarity with ASR/TTS stacks (e.g. Whisper, NeMo, wav2vec2)
- Experience with Docker, CI/CD (GitLab CI), PostgreSQL, and production ML deployment
- Strong understanding of latency (p95), WER evaluation, and conversational KPIs
- Experience in regulated environments (banking, insurance, healthcare)
- English required (French/Dutch a plus)
This is a hands-on role building real-time conversational AI systems at scale, where performance, latency, and reliability directly shape customer experience.
👉 Interested? Apply or reach out directly to
